Qwen3.8-Max-0902vsQwen3.8-Max
Across 5 shared benchmarks, Qwen3.8-Max-0902 leads overall: Qwen3.8-Max-0902 wins 5, Qwen3.8-Max wins 0, with 0 ties and an average score difference of +11.02.
Qwen3.8-Max-0902
阿里巴巴 · 2026-09-02 · Reasoning model
Qwen3.8-Max
阿里巴巴 · 2026-08-03 · Reasoning model
Qwen3.8-Max-09025 wins(100%)(0%)0 winsQwen3.8-Max
Benchmark scores
Grouped by capability, sorted by largest gap within each. 5 shared benchmarks.
Coding and Software Engineer
Qwen3.8-Max-0902 3/3| Benchmark | Qwen3.8-Max-0902 | Qwen3.8-Max | Diff |
|---|---|---|---|
| DeepSWE | 69.304 / 32极高强度思考（工具） | 56.6019 / 32极高强度思考（工具） | +12.70 |
| MLS Bench | 50.101 / 5极高强度思考（工具） | 413 / 5极高强度思考（工具） | +9.10 |
| NL2Repo-Bench | 64.901 / 12极高强度思考（工具） | 55.907 / 12极高强度思考（工具） | +9 |
AI Agent - Tool Usage
Qwen3.8-Max-0902 2/2| Benchmark | Qwen3.8-Max-0902 | Qwen3.8-Max | Diff |
|---|---|---|---|
| AutomationBench | 50.801 / 12极高强度思考（工具） | 27.309 / 12极高强度思考（工具） | +23.50 |
| Toolathlon-Verified | 73.307 / 10极高强度思考（工具） | 72.509 / 10极高强度思考（工具） | +0.80 |
Specs
| Field | Qwen3.8-Max-0902 | Qwen3.8-Max |
|---|---|---|
| Publisher | 阿里巴巴 | 阿里巴巴 |
| Release date | 2026-09-02 | 2026-08-03 |
| Model type | Reasoning model | Reasoning model |
| Architecture | MoE | MoE |
| Parameters | 2.4T | 2.4T |
| Context length | 1M | 1M |
| Max output | 131K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Qwen3.8-Max-0902 | Qwen3.8-Max |
|---|---|---|
| Text input | $2 / 1M tokens | ¥12 / 1M tokens |
| Text output | $6 / 1M tokens | ¥36 / 1M tokens |
| Cache read | $0.25 / 1M tokens | ¥1.5 / 1M tokens |
| Cache write | $2.5 / 1M tokens | Not public |
Summary
- Qwen3.8-Max-0902leads in:Coding and Software Engineer (3/3), AI Agent - Tool Usage (2/2)
On average across the 5 shared benchmarks, Qwen3.8-Max-0902 scores 11.02 higher.
Largest single-benchmark gap: AutomationBench — Qwen3.8-Max-0902 50.80 vs Qwen3.8-Max 27.30 (+23.50).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.